Journal of Veterinary Behavior (2012) 7, 103-117

POINT-COUNTERPOINT

An overview of the dog–human dyad and ethograms

within it
Paul D. McGreevya, Melissa Starlinga, N. J. Bransonb, Mia L. Cobbc,d, Debbie Calnone

a
Faculty of Veterinary Science, University of Sydney, Sydney, NSW, Australia;
b
Research Services Division, Deakin University, VIC, Australia;
c
Guide Dogs Victoria, Kew, VIC, Australia;
d
Anthrozoology Research Group, Monash University, VIC, Australia; and
e
Behaviour Counselling Service, Oakleigh South, VIC, Australia.

KEYWORDS: Abstract This article reviews the literature on the complex and variable nature of the dog–human dyad
dog–human and describes the influence of terms such as ‘‘dominance’’ on attitudes that humans have toward dogs.
interactions; It highlights a legacy of tension between ethology and psychology and notes that some practitioners
intraspecific have skills with dogs that elude the best learning theorists. Despite the widespread appeal of being able
communication; to communicate with dogs as dogs do with one another, attempting to apply the intraspecific dog etho-
interspecific gram to human–dog and dog–human interactions may have limited scope. The balance of learning the-
communication; ory and ethology on our interactions with dogs is sometimes elusive but should spur the scientific
dominance; community to examine skills deployed by the most effective humane practitioners. This process will
submission; demystify the so-called whispering techniques and permit discourse on the reasons some training
deference and handling techniques are more effective, relevant, and humane than others. This article explores
the mismatch between the use of nonverbal communication of 2 species and offers a framework for
future studies in this domain. Technologies emerging from equitation science may help to disclose con-
fusing interventions through the collar and lead and thus define effective and humane use of negative
reinforcement. The case for a validated intraspecific and interspecific canid ethogram is also made.
Ó 2012 Elsevier Inc. All rights reserved.

Introduction animals as the property of their human owners, and it

remains this way in many countries today. As science began
In the 17th century, English law viewed animals as guilty of to reveal that nature could be subject to human control, the
their actions. For example, ‘‘When in 1679 a London perception that people were vulnerable to the metaphorical
woman swung at Tyburn for bestiality, her canine partner in mystique of animals was rationalized under the banners of
crime suffered the same punishment on the same grounds.’’ zoology, taxonomy, and veterinary science. The accompa-
By the end of the 19th century, the law had changed to view nying shift in the nature of the relationship between humans
and animals that remains today is that animals became
objects of human manipulation. For a detailed review of the
Address for reprint requests and correspondence: Paul D. McGreevy,
social domination of animals in the Victorian era, we refer
BVSc, MRCVS, PhD, Faculty of Veterinary Science (B19), University of
Sydney, Sydney, NSW 2006, Australia; Tel: 161 2 9351 2810; Fax: 161 the reader to the study by Ritvo (1990).
2 9351 3957. Since the first half of the 19th century, the sentimental
E-mail: paul.mcgreevy@sydney.edu.au value of companion dogs has continued to increase. At its best,

1558-7878/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.jveb.2011.06.001
104 Journal of Veterinary Behavior, Vol 7, No 2, March/April 2012

the dog–human relationship is characterized by strong at- members of a social group (Petherick, 2010). Social organi-
tachment and the optimal well-being of both dogs (Kotrschal zation in Canis familiaris can be studied by observing the
et al., 2009) and humans (Barker et al., 2003). At its worst, the way in which dogs gain access to and retain resources
dog–human relationship is associated with animal abuse (for a (Drews, 1993). Dominance is characterized as an aspect
comprehensive review, see Ascione, 2008) and the routine of a relationship between 2 or more animals in a social
destruction of unwanted and abandoned dogs (McGreevy grouping rather than an attribute or trait of an individual.
and Bennett, 2010). Canine behavior problems can have an The ‘‘dominant’’ animal is considered to have higher status
adverse effect on the well-being of owners and on the wider over another or others in the group (Petherick, 2010). A
community (Voith, 2009), whereas the abandonment or loss dominance relationship involves a simultaneous expression
of companion animals can also be associated with human psy- of both dominance and submission (Schenkel, 1967). A
chopathology (Hunt et al., 2008). For these reasons, harmony dominance relationship can be established without any
between the 2 species is of critical importance. sign of aggression only indicated by a submissive or
To better understand canine social behavior and, in turn, appeasing posture from one of the protagonists. An appeas-
the dog–human dyad, we do well to first study the ways in ing attitude from the dog toward humans may therefore
which social harmony is created and maintained in stable indicate a man–dog dominance relationship. An individual
groups of dogs. The peacefulness that usually defines such dog’s motivation to gain access to a particular resource
established dog communities reminds us that there are very may be subject to some flux (Bradshaw et al., 2009)
few breaches of social order and that aggression is rare and resource-holding potential may be context-specific
(Bradshaw et al., 2009). This is underpinned by clear signal- (Shepherd, 2002), but this should not demean its impor-
ing and deference, delivered not demanded (Overall, 1997). tance. The ability to learn is similarly context-specific
Rough physical contact is far more often part of play than (McGreevy and Boakes, 2007). It is clear that dogs view
of violence, and it is foreshadowed by strong signals humans differently from the way they view other dogs
(Horowitz, 2009a). (e.g., Rooney et al., 2000). Nevertheless, we would do
This article will review the literature relating to our well to study the role of intraspecific canine social interac-
current understanding of the complex nature of the dog– tions in dog–human interactions and human–dog interac-
human dyad and examine its characteristics. First, it con- tions, if for no other reason than that humans are
siders the origins of social-domination belief systems and regularly bitten by dogs defending certain resources. Ag-
how social order is usually studied. Second, it reviews canine gression is defined as deliberate threat and/or attack compo-
social behavior and social learning. Third, it examines nents of agonistic behavior with potential to cause injury
canine–human interactions, examining the extent to which associated with conflict and competition (Brain, 2010).
they reflect or are informed by the canine ethogram and by Recent data suggest that higher scores for owner-directed
learning theory. We summarize by exploring our ability to aggression are associated with male dogs and female
apply the canine social ethogram to handling and training owners (Hsu and Sun, 2010), but this does not necessarily
and to assess the limitations of this approach, and we suggest indicate social dominance as a cause. It is critical that the
that it is possible to estimate the contribution of ethological role of testosterone in impulsive and reactive aggression
and psychological principles in the manifestations of certain among companion dogs is better understood. And only
responses. However, the emphasis here is on estimation, we when we study how aggression and successful affiliative
are not suggesting that we can quantify the absolute roles of activities (such as play) (Horowitz, 2009b) emerge in
learned rather than innate responses. Context can determine dog–dog dyads, will the subtleties of dog body language
whether learning theory will have a greater influence or be and our ability to offend our canine companions, however
more informative than ethology when training and handling inadvertently, become clear.
animals (McGreevy et al., 2009); thus, we avoid using cues Social order helps animals within a social group learn
of ethological significance if they run counter to a given which of them can defend resources and displace another
training outcome. For example, because it so reliably trig- from them (McGreevy, 2004). This learning is underpinned
gers playful responses rather than conditioned responses, by communication and effectively reduces aggression
play-bowing (an innate canine meta-signal for play) is rarely (McGreevy, 2009). It can exist without the necessity for
used as a visual discriminative stimulus in training. Using individuals to have a sense or concept of their own status.
this approach, we offer a framework that helps to describe Humans have gone to great lengths to interpret animal
this effect in dog–human interactions. behavior in terms of social order. Some of the theoretical
constructs that have arisen in the process, such as ‘‘domi-
nance,’’ have acted as obstacles to successfully extending
Origins and implications of the social dominance our understanding of animal behavior (Friedman and
belief system Brinker, 2001). Many authors (e.g., Semyonova, 2003) crit-
icize labels such as ‘‘dominance’’ because they evoke
Social order can be understood as the product of emotional responses in the observer and can prejudice
dynamic and situation-specific relationships between interpretations, interfere with verifiability of behavioral
McGreevy et al Overview of the dog–human dyad 105

responses, and are tightly bound by the human perspective. challenged (reviewed by [Bradshaw et al., 2009;
It is important to note here that social dominance order (or Semyonova, 2003; van Kerkhove, 2004a; 2004b; Steinker,
orientation) is a term used in psychology to describe a 2007]). The merits of the putative linear hierarchy of wolf so-
personality variable that predicts social and political atti- cial order as an appropriate model to apply to the social be-
tudes. It is distasteful to many because it reflects an individ- havior of domestic dogs has been widely discussed (van
ual’s preference for inequality among social groups (Pratto Kerkhove, 2004a; 2004b; Scott and Fuller, 1965; Serpell,
et al., 1994) and may in some ways relate to right-wing 1995; Bradshaw et al., 2009; Steinker, 2007).
authoritarianism. Mech (1999) described 13 years of studying the social
Non-Western philosophical frameworks, such as order within wild wolf packs, and the results of this
Buddhism, believe that animals exist in their own right, research contradict most of the widely held beliefs regard-
rather than as property, as they are represented in Western ing dominance hierarchies in this species that have been
law. Thus, it is pertinent to assess the validity of the assumed to be applicable to the domestic dog. He found
dominance construct in terms of Western culture’s sanction that family groups, rather than a linear hierarchy, were
of the control of man over nature (Misra, 1995). observed in wild wolf populations. Based on these results,
In addition to the effect of cultural attitudes, it is also van Kerkhove (2004a; 2004b) and Yin (2009) have sug-
worth considering the potential effect of inattentional gested that ‘‘wolf pack theory’’ does not apply to domestic
blindness on human perception of dog behavior. Inatten- dogs and challenged the idea that humans should maintain
tional blindness is the notion that ‘‘we rarely see what we social stability in their interactions with dogs by adopting a
are looking at unless our attention is directed to it’’ (Mack, ‘‘top-dog’’ role. Furthermore, Coppinger (2001) has argued
2003). This concept highlights the intimate link between that dogs are not pack animals. That said, there is evidence
our perception and our attention. Steinker (2007) argued that, where resources are sufficient, large groups of dogs
that by labeling a dog as ‘‘dominant,’’ the humans involved can occupy a single area with minimal conflict (Bradshaw
begin to interpret many behaviors as evidence of ‘‘domi- et al., 2009) and that dogs can cooperate in tasks
nance’’ and ignore any evidence to the contrary. The con- (McGreevy, 2009).
cept of inattentional blindness may further assist with Scott and Fuller (1965) believed that social hierarchies
understanding why people are very likely to see, remember, effectively reduce and minimize overt aggression between
and interpret dog–dog and dog–human interaction in terms individual members of the pack but noted that a linear hier-
of the dominance framework with which they are familiar. archy does not seem to be a factor for the domestic dog in
This concept illustrates the power of our intentions in deter- maintaining social harmony. They evaluated social order in
mining what we see and what we do not (Mack, 2003). their study using a ‘‘dominance test’’ on puppies at 5, 11,
On a practical level, it is clear that some practitioners and 15 weeks of age. Two puppies were placed in a pen
have skills with dogs that elude the best learning theorists; with a bone for 10 minutes. Dominance was defined as a
they may insist that they are using canine ethology and condition in which 1 puppy kept possession of the bone
imposing canine social order to communicate with dogs. As for at least 8 out of 10 minutes. The dogs were rated as
a result, the role of dog owners and handlers as leaders, ‘‘dominant,’’ ‘‘incompletely dominant,’’ and ‘‘subordinate,’’
alphas, and trainers is widely debated. A popular view that and the effect of dominance on the amount of fighting was
has, until recently, prevailed in dog-training circles is that assessed. The observations varied across 3 breeds: Fox
dog–dog interactions mirror wolf–wolf interactions. An terrier, Basenji, and Shetland sheepdogs. Breed differences
example is provided by Bauer and Smuts (2007), who were found in the association between ‘‘dominance’’
stated that ‘‘as studies of captive wolves found that postural (defined as control over a bone) and an effective control
asymmetries were consistently unidirectional within dyads system over fights between individuals. Thus, the notion
and that dominance hierarchies based on these postures that dominance order could control fighting was upheld in
showed a high degree of linearity . such postures were 2 of 3 breeds of dogs when observed between 5 and
therefore considered reliable indicators of dominance for 15 weeks of age. However, if we place these findings in
our purposes’’ (in studying companion dogs). These authors the context of the difficulty other scientists face when
then went on to draw conclusions about play behavior in attempting to develop tests in puppies that predict their
companion dogs on the basis of their relative social status, behavior as adults (Wilsson and Sundgren, 1998; Batt
which were based on wolf data. et al., 2009), we can only speculate how any association
Bradshaw et al. (2009) reviewed data available on social between dominance and fighting might vary in dogs of
hierarchy in the dog from the perspective of dominance other ages, let alone in other breeds.
hierarchy. It seems that, historically, the suggestion that Scott and Fuller (1965) described 9 fundamental behav-
humans could and should adopt the role of pack leader was ioral systems for dogs (investigative behavior, epimeletic
based on 2 implicit assumptions: first, wolves are the ances- behavior, et-epimeletic behavior, allelomimetic behavior,
tors of domestic dogs, and second, that a linear hierarchy agonistic behavior, sexual behavior, eliminative behavior,
exists in wolf packs (van Kerkhove, 2004a; 2004b). Although ingestive behavior, and comfort-seeking behavior [shelter-
the first point is widely accepted, the second has been seeking]). They concluded that not only were these relevant
106 Journal of Veterinary Behavior, Vol 7, No 2, March/April 2012

to dogs of all breeds, they were also relevant as a frame- Studying domestic canid social order
work for comparing human and dog behavior patterns.
While acknowledging that, in many ways, dogs and humans Bradshaw et al. (2009) posit that resource holding poten-
are different in terms of anatomy, physiology, and behavior, tial (Parker, 1974) may be less applicable to dogs than to
these authors noted that social behavior patterns are similar other species, but that nevertheless it is useful because it
enough, in many contexts, to be mutually recognizable. offers the concept of subjective resource value as a factor
A brief consideration of analogues of dog–dog interac- influencing the escalation of conflicts. Bradshaw et al.
tions that arise in human–dog dyads suggests that there are (2009) propose that the subjective resource value, in com-
at least some human–dog interactions that align with the bination with associative learning, explains antagonistic
canid intraspecific social ethogram. Analogues, such as encounters between dogs more simply than traditional
allogrooming, can be useful for humans needing to groom dominance theory. This is a useful contribution to explain
dogs (McGreevy et al., 2005), whereas others may raise resource-related aggression but it fails to either acknowl-
challenges because there can be tension between ethologi- edge the possible role of personality dimensions or explain
cal and psychological constructs in training. For example, the mechanisms of dog–dog antagonistic interactions in the
although gaze-averting is a deference display in dogs absence of clearly disputed resources.
(Vas et al., 2005), during some training sessions, it is It could be argued that encounters between 2 members
critical to keep a dog’s attention focused on the handler. of a dyad are never resource-neutral. Fighting may be
Furthermore, breed differences in attention to human cues exhibited in different contexts, including competition over
have been demonstrated (Gácsi et al., 2009a), and differ- a resource, and contexts not clearly related to resource
ences in aloofness (McGreevy, 2007a) and even retinal possession (Hahn and Wright, 1998, cited by Wright,
anatomy (McGreevy et al., 2004) may account for the 2004). For example, when dogs first meet, one of them is
lack of attention some dogs pay their handlers. more the territory holder than the other (even if only by
Vas et al. (2005) developed a scale for assessing behav- virtue of being on that site for longer than the protagonist)
ioral responses of dogs to approaches by unfamiliar and perhaps presence on the territory is a quasi-resource.
humans. These authors note that humans may have selected But when meeting for the first time, how can dogs be
dogs based, inter alia, on variation in monitoring of and sure they are valuing or fighting over the same resource?
response to human cues. Certainly, this could be a first Before any associative learning about their relationship
step in selecting dogs that may excel at protection, guard- can take place, their behavior may represent a manifesta-
ing, and herding, all of which share related behaviors. tion of positive and negative personality dimensions such
Dogs have been selected for adaptations to human social as those described by Sheppard and Mills (2002) or the
life, and these adaptations have led to marked changes in 5 personality factors described by people interviewed in
their communicative, social, cooperative, and attachment the study by Ley et al. (2007).
behaviors toward humans. In a review of canine social A given dog’s ability to impose social status at the first
cognition, Miklósi et al. (2004) state that through a complex meeting would seem to lay the platform for future interac-
evolutionary process, dogs became adapted for living in tions over resources. For example, an extroverted dog may
human society. Therefore, the human environment and social set the stage for subsequent encounters with an introverted
setting now represents a natural ecological niche for this dog over equally valued resources. Some dogs may be
species. Dogs are extremely flexible in how they process unable to interpret the intentions of unfamiliar dogs when
spatial information and can simultaneously use cues from they approach conspecifics; they may be less concerned
different sources and rank the cues based on the complexity with resources than with the need to obviate a perceived
of the environment (Fiset et al., 2000, 2006; Fiset and threat. Clinical experience suggests that some dogs will
LeBlanc, 2007). If humans are part of that environment, we behave in an aggressive manner toward any unfamiliar dog,
must consider that dogs trained to work in the presence of in any setting, at considerable distances (Debbie Calnon,
humans who issue discriminative stimuli may disregard personal communication). For these dogs, it is very difficult
information that they, themselves, collected (Szetei et al., to identify the value of a resource because the most likely
2003). Given that companion dogs behaved differently primary motivation appears to be fear or anxiety. This
(attempted a forbidden task) when the owners were in the prompts us to ask whether access to self-defense mecha-
room compared with when they were not in the room nisms (and actions taken to reduce the perceived risk of
(Schwab and Huber, 2006; Horowitz, 2009b), it seems likely harm to self) is the resource these dogs value.
that intimate human–dog relationships (as occur in compan- Scientists generally base their measurements of social
ion dogs) may predispose dogs to behave in a socially hierarchies on who displaces whom from food and, less
dependent manner. All of this suggests that canine scientists often, on who initiates contact with whom. However, it is
seeking to advance communication in the dog–human dyad important to ask whether the order that predicts displace-
must grant ethologically relevant mechanisms as much ment and appeasement is sometimes based on relative value
attention as mechanisms that align with learning theory of resources and sometimes on fear. Regardless, the ques-
(McGreevy and Boakes, 2007). tion is whether such social order among dogs can include
McGreevy et al Overview of the dog–human dyad 107

humans and whether perceived breaches of order may of the work on cognition in dogs has focused on testing pu-
explain how humans occasionally get bitten. It is possible tative effects of domestication of pet dogs and comparing the
that inconsistency on the part of humans can create results with captive, usually hand-reared wolves. Wolves in
behavioral conflict in nearby dogs and the resultant frus- these populations do not respond to human cues about
tration can trigger aggression. Dogs that bite humans are location of items in the same way that dogs do, leading to
usually, but not always, thought to bite because of fear and the conclusion that ‘‘pointing’’ or ‘‘showing’’ are behaviors
anxiety (McGreevy and Calnon, 2010). Of course, all non- developed because of close contact with humans, possibly
biting dogs are not necessarily free of fear or anxiety solely because of domestication (Miklosi, 2007).
because they are sure of their social status. Nevertheless, The ability of dogs to gain access to particular resources,
care is warranted in human activities that may amount to retain their own resources, and displace other dogs from
ethologically relevant social threats to dogs. resources is a critical element of social order. Dogs are
Conflicts between dogs living in the same household are limited to burying prized objects, but humans can stash
most often between members of the same sex, and more resources in pockets, boxes, and cupboards that only they
often involve females than males (Sherman et al., 1996; can access. Dogs do not feed adult conspecifics nor, for that
Wrubel et al., 2011). That said, the triggers for these ag- matter, do they dictate when they take exercise. A dog’s
gressive encounters are generally reported by dog owners behavior can be manipulated by identifying resources
who are untrained in making behavioral observations. valued by the individual dog. Resources such as food and
Thus, it is difficult to identify why these conflicts occur exercise can be used by humans to encourage and discour-
and why they are more prevalent between members of the age particular canine behaviors, illustrating one distinct
same sex than between members of the opposite sex. difference between intra- and interspecific social relation-
Free-ranging dogs living in groups are reported to show a ships. It follows that resource supply, resource guarding,
linear hierarchy, but although there are differences in fre- and resource-related frustration (on the part of dogs) can
quency of agonistic interactions between males and fe- prove problematic in some dog–human dyads.
males, there is no clear indication whether there is a Most operant conditioning uses learning theory to modify
separate hierarchy for each sex (Pal et al. 1998). Scott dog behavior. We control access to the resources and, using
and Fuller (1965) found that when male and female puppies them, can train all the behaviors we regard as desirable. This is
contested a resource, males tended to win. They postulated something other dogs do not appear to do as effectively as we
that this was because males were typically larger. However, do. Thus, again, we need to be cautious because it may be
between sexes, size had no effect on the outcome of con- wrong to assume that dogs want to control other dogs through
tests in female–female pairs, and only a weak effect in access to resources. From ethological descriptors, a social
male–male pairs. Scott and Fuller (1965) concluded that a animal is accorded rank through its ability to gain access to
relationship tends to reflect the differential capacities of specific resources. However, we emphasize that it does not
the 2 individuals involved. Thus, dogs of different sex follow that the animal is motivated to take control of resources
tend to have more defined relationships, with one member to attain rank. In other words, the animal is not motivated by a
typically playing the dominant role and the other playing desire to be dominant for the sake of attaining rank per se.
the subordinate, whereas dogs of the same sex tend to
have relationships that are less well-defined, with dominant The role of social learning
and subordinate roles switching readily. The latter relation-
ship is more likely to trigger aggressive behavior generated Social learning occurs when an individual learns by
by conflict. observation of another individual (Ligout, 2010). In some
species, such as domestic chickens, the social status of
The relevance of canine social behavior the demonstrator has a strong influence on the perceived
value of the information it imparts (Nicol and Pope,
Moehlman (1987) reminds us that the regulation of 1999). Social learning is an important aspect of the dog’s
social structure and behavior in wild canids reflects charac- social behavior (Horowitz, 2009a); thus, we should con-
teristics of the canid (size, weight, and sex), the group sider how canine ethology informs the way in which we
(group size, territory, and reproductive strategies), and apply learning theory.
access to food (temporal and spatial distribution of prey). Cognitive tests can include truly novel components that
Domestication has changed many of these variables beyond require learning during the test, including situations in
recognition, and perhaps most of all in the supply of which dogs learn from watching other dogs successfully
resources, especially food. perform and be rewarded for performing a novel task
Homologous behavioral mechanisms can be identified (Range et al., 2009). It has been shown that the pups of
between wolves and dogs, but the best model for describing trained drug-detection bitches learned to pay attention to
social relationships among domesticated dogs reared in a target odors from watching their mothers (Slabbert and
home environment derives from that environment rather Rasa, 1997). In another study, observer dogs were able to
than any wild canid social structure (Wright, 2004). Much adjust their search behavior for hidden food depending on
108 Journal of Veterinary Behavior, Vol 7, No 2, March/April 2012

the knowledge gained by observing and interacting with a Scott and Fuller (1965) described the complexity of
conspecific (Heberlein and Turner, 2009). It has also been dog–human interactions and noted that these are variable
shown that subordinate dogs, as defined by the balance of and capable of change. They identified 21 possible different
leadership and deferential behavior, learned a detour task types of social relationships between humans and all the
more quickly than dominant dogs if they observed it dem- breeds of dogs. Because several domestic dogs are sexually
onstrated by another dog, but there was no difference if the neutered, we added the extra categories of neutered male
demonstrator was human (Pongrácz et al., 2008). In a study and neutered female to the original table from Scott and
of 118 dogs split into 2 groups, the group that watched a Fuller (1965). The amended version of this table is repro-
human manipulate a test/treat box was faster in successfully duced later in the text (Table 1) and shows 27 possible
opening the box and spent more time interacting with it different types of social relationships between dogs and
than the group that did not (Marshall-Pescini et al., humans.
2008). Thus, we should consider the extent to which dogs It seems that, within contemporary households, dogs
may learn from us with no intended operant conditioning. may have negligible, frequent, or irregular interactions with
The prospect of being able to model behaviors for dogs other dogs or people (McGreevy, 2009). Thus, the applica-
to adopt mimicry is beguiling and is strongly hinted at by bility of canine ethograms may be limited for some dogs
studies of model-rival (McKinley and Young, 2003), and living with humans. Indeed, dogs living in single-dog
possibly also the effect of pointing and referential gazing households may not even become fluent in their own lan-
(Elgier et al., 2009; Gácsi et al., 2009a,b). guage. Whether this has an effect on their fluency in com-
municating with both conspecifics and heterospecifics is
Canine–human interactions worth considering. Evidence of the adaptability and flexi-
bility that dogs have when living and communicating
Among the challenges to understanding the way dogs fit with other species was studied by Feuerstein and Terkel
in or not to human households are the variability of dog- (2008), who found that a first encounter taking place at
keeping styles (Masters and McGreevy, 2008) and human an early age (up to 6 months in cats and up to 1 year in
personality types (McCrae and John, 1992). Households dogs) enabled most cohabiting dogs and cats to appropri-
can be occupied by single or multiple dogs just as they ately interpret the particular body language displayed by
can be by single or multiple humans (McGreevy and the other, even when the signal had an opposite meaning
Masters, 2008). for both species. Furthermore, they showed that the youn-
In what remains the largest dog behavior research project ger the animal at first encounter, the better this understand-
of its kind, Scott and Fuller (1965) acknowledged that the ing was, and the more chance there was of establishing an
chief relationships they studied were the human males and amicable relationship.
females with young dogs. The gender of human participants Social groups in feral dog contexts are arguably subject
in dog–human dyads is of fundamental importance. Although to less flux than those in the human–dog domain. Feral
we know that dogs’ cortisol concentrations often reduce dogs do not meet strangers in their den on a regular basis,
with human interaction in general (Hennessy et al., 2001; visit parks, or go on holidays. Dogs have not evolved to
Tuber et al., 1996), there is evidence of a more marked know that the new social groups that arise in the company
reduction in cortisol responses in shelter dogs when being of strangers, on a visit to a park, or a trip to a holiday
stroked by women (Hennessy et al., 1998). destination, are not going to last for the rest of time.

Table 1 Fundamental classification of social relationships of dogs and humans

Dog Human
Type Male (MD) Female (FD) Young (YD) Male (MH) Female (FH) Young (YH)
Dog
Male MD–MD
Male neuter (MN) MN–MD MN–FD MN–YD
Female FD–MD FD–FD FD–YD
Female Neuter (FN) FN–MD FN–FD FN–YD
Young YD–MD YD–FD
Human
Male MH–MD MH–FD MH–YD MH–MH
Female FH–MD FH–FD FH–YD FH–MH FH–FH
Young YH–MD YH–FD YH–YD YH–MH YH–FH YH–YH

Adapted from Scott and Fuller, 1965, Genetics and the social behavior of the dog, University of Chicago Press, Chicago, IL. Ó 1965 The University of
Chicago. All Rights Reserved.
McGreevy et al Overview of the dog–human dyad 109

Making sense of how these novel groupings will affect  Restraint and (giving or receiving) aversive stimuli
access to resources either relies on learning gradually about (Table 2).
each resource and who is allowed it and when, or it  Tactile activities (allogrooming and resting) (Table 3).
involves some social order, perhaps based entirely on  Meeting unfamiliar individuals (Table 4).
deference, that removes the need for constant disputes.  Sharing resources/playing with objects (Table 5).
There seems to be merit in working out swiftly and  Greeting familiar individuals and playing without
painlessly who must voluntarily defer to whom. Dogs in objects (Table 6).
the human domain may become highly skilled in respond-  Nontactile interactions (Table 7).
ing appropriately to this cognitive challenge.
Notwithstanding the variability described earlier in the The instances in which dogs do not react to humans as
text, it is possible to map out common dog–human and described in Table 2 are usually because of the relative
human–dog interactions by following the framework for difference in height. Common human–dog interactions
exploring horse–human interactions offered by McGreevy in this domain that have no analogue in the canine
et al. (2009). This is discussed later in our concluding ethogram include the use of collars, chains, harnesses,
suggestions for further research. However, we need to be headcollars, catch poles, lifting dogs off the ground, and
cautious with this approach. Lit et al. (2010) investigated confinement.
owner reports of dog–human interactions and showed Table 3 shows that dogs will attempt to use almost all
how complex owners’ interpretations of such interactions their repertoire of tactile activities with familiar humans.
can be. Elements of the dog–dog social ethogram may look However, common human–dog interactions that have no
similar to what humans see when dogs and humans interact analogue in the canine ethogram include scratching the
but, so far, we cannot be certain that they look similar chest, putting on collar/harness, combing, brushing, nail-
to the dog. We must be aware that dogs may not see or inter- clipping, expressing anal sacs, opening mouth, administering
pret these categories in the same way we do. Nevertheless, oral medications, applying topical medications, dressing
there is merit in offering a framework that has its basis wounds, toweling dry, hands-on training, bathing and
in dog–dog interactions because it may explain where cleaning ears, patting as opposed to stroking or scratching,
errors in human attempts to communicate with dogs most clipping of coat, kicking, pushing, and smacking.
likely occur. Table 4 shows that, where our bipedalism does not
Using our own observations and pooled experience, and interfere, dogs generally attempt to use almost all their
acknowledging the importance of breed differences in repertoire of greeting activities with familiar humans.
signaling (Goodwin et al., 1997), we have tabulated the However, our ability to mimic canine responses is severely
ways in which dogs interact with each other and with limited. Furthermore, common human–dog interactions
us, and have contrasted these interactions with some of that have no analogue in the canine ethogram include hug-
the activities we impose on dogs using the following ging familiar group members, patting on head, ignoring,
6 domains: screaming, and running away.

Table 2 A consideration of restraint and (giving or receiving) aversive stimuli as they arise in dog–dog, dog–human, and human–dog
interactions

Dog–dog behavior Possible human–dog equivalent (without apparatus) Dog–human analogue of dog–dog behavior
Attempt to escape restraint Present Present
Bite Absent Present
Body block Present Presenta
Boxing Present Present
Grip with mouth/teeth Present: hold a body part firmly with hand Present
and fingernails
Lunge Present: lunge forward with hands outstretched Present
towards dog
Pin with chest/body weight Possible: pin with hands Absentddue to relative difference in height
Pin with muzzle Possible: pin with hands Absentddue to relative difference in height
Snap Possible: quick lunge with one or both hands Present
without contact
Stand over Possible: may do so naturally due to relative height Absentddue to relative difference in height
Threaten to bite as a form Absent Present
of body blocking
a
Impeded by human bipedalism.
110 Journal of Veterinary Behavior, Vol 7, No 2, March/April 2012

Table 3 A consideration of tactile activities (allogrooming and resting) in dog–dog, dog–human, and human–dog interactions
Dog–dog behavior Human–dog equivalent Dog–human response?
Grooming nibble Present: brush/scratch Present
Lick Present: kiss/spongeing/wiping Present
Mouth Present: isolation and restraint of particular body part Present
Nuzzle face/ears Present: massage face/ears with fingers Present
Resting in physical contact Present Present

There are some compelling analogues in Table 5. How- ethogram in interactions with humans and where the limits
ever, common human–dog interactions that do not align to these attempts and to reciprocation generally lie. If we
with the canine ethogram include throwing objects, not consider contexts in which dogs and humans interact, we
chasing objects of value (e.g., thrown articles), giving can make predictions about the outcome of the interactions
food (including bones, chews, titbits), giving play objects, based on the relative role of learning theory and the canid
and kicking balls. intraspecific social ethogram (see Table 8).
Again, Table 6 shows that humans struggle to recipro- Table 8 shows how regularly the most common human
cate in ways that align with the canine ethogram when physical interactions (restraint, brushing and combing,
playing without objects. In addition, common human–dog training, feeding, watering, and veterinary care) have no
interactions that have no analogue in the canine ethogram analogue in the canine social ethogram and how patchy is
include some of the most popular dog sports: off-lead train- the human’s ability to offer plausible analogues of dog–dog
ing, including agility, canine freestyle, and tricks. It is interactions. Taken together, Tables 2-8 show the many
worth noting here that, among dogs, it has been found ways in which dogs may attempt to use their social skills
that more advantaged individuals do not consistently relin- on humans, where their height permits and our bipedalism
quish their advantage to facilitate play (Bauer and Smuts, does not obstruct such attempts. But they also seem to
2007). Role reversals do occur, but certain social conven- demonstrate that, when we consider dog training and
tions influence which behaviors could be used during role handling, there are limitations to the usefulness of both the
reversals (Bauer and Smuts, 2007). canid ethogram and learning theory. These limitations are
In Table 7, the imbalance between the 2 species appears discussed later in the text but, taken together, our shortcom-
most striking. There are numerous examples of how ings in the use of the canid ethogram to communicate with
humans generally struggle to use elements of the canine dogs should remind us that successful dog handling is
ethogram to communicate with dogs. In addition, there not successful dog mimicry. However, by the very same
are common human–dog interactions with no analogue in token, we need to acknowledge that these limitations
the canine ethogram: blowing or whistling in the face, wav- can help to explain instances of dogs biting humans in
ing arms around, hand signals, and gesturing (not pointing) the absence of fear or anxiety. For example, making the
with hands or head. wrong move when a dog has placed its forepaws on
This examination of intraspecific interactions shows how one’s shoulder while staring you in the face can trigger a
commonly dogs seem to deploy elements of their social bite (McGreevy, 2009).

Table 4 A consideration of meeting and greeting activities among unfamiliar individuals in dog–dog, dog–human, and human–dog
dyads

Dog–dog behavior Human–dog equivalent Dog–human response

Body block Present Present
Circle Possible Present
Genital sniffing/licking Absent Present
Increased postural tonus Present: stand tall, tension in body Present
Lift paw onto forequarter Present: place hand on shoulder Presenta
Lunge Present: lunge forward with hands outstretched toward dog Present
Mount Possible: when lifting a medium-sized dog Presenta
Raise hackles Absent Present
Snap Absent Present
Sniff under tail Absent Presenta
Touch noses Present: not as easily as in dogs. Presenta
a
Impeded by human bipedalism.
McGreevy et al Overview of the dog–human dyad 111

That said, one of the key aspects of evolving with humans

Table 5 A consideration of the ways in which dog–dog,
dog–human, and human–dog dyads may share resources is that dogs have learned to coexist with people, and the
or play with objects fact that dogs have been so successfully used by people
demonstrates that an effective means of communication
Dog–dog Dog–human does exist.
behavior Human–dog equivalent response The debate between the relative roles of learning and the
Nose object Present: push object Present ethogram mirrors the historic tension between ethology and
with hand psychology. On one hand, we have the suggestion that
Tug Present Present every response made by a dog (including aggression) has to
be learned and, on the other hand, we have practitioners
proposing that they only have to mimic wolf behavior to
Applying the canid ethogram to dog–human gain compliance and a bond with dogs. The middle ground
dyads suggests that dogs are likely to rely on the social repertoire
they have evolved to use with other dogs unless they have
Aspects of our body language and behavior may stim- been socialized with other species.
ulate resource guarding and different forms of anxiety in
dogs. Human activities that displace a dog may culminate
Communication modalities used in training
in a bite if the dog has learned to defend its resources. Dogs
that have been granted free access to some resources as
In general, we train dogs with negative reinforcement
juveniles learn to defend these and, quite plausibly, others.
(NR), positive reinforcement, and punishment. Applying
Thus, the value of one resource may have implications for
physical pressure-release (NR) and providing reinforcers
the unexpected defense of others. It is easy to see how this
only when a desired behavior is performed (positive
could lead to unanticipated displays of aggression and to a
reinforcement) are seldom reported in the canine intra-
dog being labeled dominant.
specific ethogram, so it seems that we generally train dogs
Our relative height means that dogs automatically look
with modalities that have minimal inherent relevance to
up to us and that this may lend us an ethological advantage.
their social learning. This may be a significant failing
It is possible that this makes us ‘‘super-dogs’’: initiators of
on our part.
expeditions, exercise, grooming, play, and feeding, leaders
To be effective and safe in interactions with dogs, our
who would never be worth questioning, but clearly this is
actions outside operant conditioning rely less on timing and
simplistic. The evidence presented by Bradshaw et al.
consistency and more on being able to do the following:
(2009) suggests that there is no leader within the dog–dog
domain, so consideration should be given as to how dogs  Interpret dogs’ body language and assess their motiva-
could transfer such a role to another species. However, tion correctly;
the absence of a clear leader in a study of 19 neutered  Mimic appropriate elements of the ethogram (e.g., play-
male dogs presented by Bradshaw et al. (2009) does not bowing), notwithstanding their context-specificity;
imply that humans cannot become integrated into social  Avoid threatening dogs, their young, and their resources;
constructs that have their origins in the social ethology of  Neither inadvertently nor consistently defer and thus
C familiaris. release resources that are valued by the dog;
Whether dogs have evolved to use their skills for social  Offer appropriate models that may be of use in social
order with another species (most notably us) is contentious. learning;

Table 6 A consideration of the ways in which dog–dog, dog–human, and human–dog dyads greet familiar individuals and play
without objects
Dog–dog behavior Human–dog equivalent Dog–human response
Chase Present Present
Jaw sparring Present: tapping either side of jaw with hands Absent: due to relative
while dog tries to catch hands difference in height
Mount Possible: when lifting a medium-sized dog Present
Mouth at legs or neck/face Absent Presenta
Paw Present: reach out with hand Present
Pounce Present: lunge Present
Rubbing Present Present
Wrestling while running Absent Present
a
Impeded by human bipedalism.
112 Journal of Veterinary Behavior, Vol 7, No 2, March/April 2012

Table 7 A consideration of the ways in which dog–dog, dog–human, and human–dog dyads may communicate using nontactile
mechanisms
Dog–dog behavior Human–dog equivalent Dog–human response?
Avoidance Present Present
Bare teeth Present: smile Present
Bark Present: shout Present
Body shake Absent Present
Excitement bark Present: shout Present
Frustration bark Present: shout Present
Warning growl Present: can be imitated with low frequency tone of voice Present
Head and neck roll Absent Present
Lick lips Present Present
Look away Present: avert eyes Present
Pilo-erection Absent Present
Play bow Present: can be imitated Presenta
Play growl Possible: can be imitated with higher frequency tone of voice Present
Prance Possible: can be imitated Presenta
Relaxed gaze into face Present Present
Stalk Present: can be imitated Presenta
Stare Present Present
Submission grin Present: smile Present
Submissive posture, e.g., inverted Present: crouch Presenta
U-position
Tail wag Absent Present
Whine Present: can be imitated with high frequency Present
tones and use of Motherese
Postural tonus, e.g., U-position, stiff Present: when tense or confident Presenta
legs and movement
a
Impeded by bipedalism.

 Provide useful information (e.g., by pointing at objects universal interpretation of actions that accompany agonistic
and caches of interest). responses in both species: fixed stares, stalking, and high
postural tonus. It may be that dogs learn more quickly when
These seem to be the very qualities that natural dogfolk hand signals are used in training rather than words (Soproni
(and possibly many so-called dog whisperers) possess. et al., 2001, 2002). This may support the premise that
Capturing, defining, and measuring these qualities and handlers who process ideas or communicate pictorially
training less gifted handlers to reproduce them may be rather than verbally have made a big leap in their effort
the enduring legacy of the dog whisperer phenomenon. to learn ‘‘dog language,’’ dogs being a visually oriented
Some dogs and some humans seem to have developed (rather than a verbally oriented) species.
their own fluent means of communication. This involves an Yin (2002) described the subtypes of barking used by dogs
ability to read body language, perhaps underpinned by a in different contexts, but many of the subtleties of canine

Table 8 A conceptual tabulation of the relative roles of learning theory, the value of resources, and the canid intraspecific social
ethogram in various common contexts

Relative role of learning The value of the focal

Alignment with the canid theory ‘‘How readily can resource (V), in combination
Context intraspecific social ethogram responses be conditioned?’’ with associative learning
Human taking resources from dog 11111 1 1111
Dog taking resources from human 11111 111 1111
Dog grooming human 11111 1 1
Human grooming dog 1 111 1
Hands-off play and training 11 1111 1111
Hands-on play and training 1 111111 1
Veterinary activities – 1 –
McGreevy et al Overview of the dog–human dyad 113

vocalization remain unknown. Acoustic signals, such as discriminate between types of loud vocalizations, such as
growls, barks, and whines, have a role in canine communica- those that accompany human anger.
tion. Growls intended as a warning are of a lower frequency Handlers may also rely on auditory cues from their dogs.
than growls in play, and are longer, but growls in play have In a study of Hungarian herding dogs (Mudis), listeners,
lower formant frequency dispersions than aggressive regardless of their experience with dogs, were able to
growls (Faragó et al., 2010). Formant frequency dispersion categorize bark situations in a way that differed signifi-
is a measure influenced by the length of the vocal tract; cantly from that expected by chance alone (Pongrácz et al.,
thus, a low formant frequency dispersion is a reliable 2005). Associations were strong for particular bark samples
indicator of a larger animal, and a high formant frequency correlated with peak and fundamental frequency and inter-
dispersion indicates the sound is coming from a smaller bark intervals.
animal (Fitch, 1997; Taylor and Reby, 2010). Barks that sig-
nal aggression also have a lower frequency than barks that Relationships between dogs and their handlers
signal play or ‘‘happiness,’’ and have shorter inter-bark inter-
vals (Pongrácz et al., 2006). Growls associated with food- A focus on mechanisms that work in dog training is
guarding appear to have a strong deterrent effect on other useful but may obscure the importance of the relationships
dogs taking a meaty bone, but warning growls associated between dogs and their handlers. This possibility has been
with a threatening stranger are also low in frequency and studied in military working dog contexts. Lefebvre et al.
have a weaker deterrent effect on dogs taking a meaty bone (2007) found that the more attention the dog received
(Faragó et al., 2010). This would suggest there are elements from the handler, even if much of this was passive and
of either context or fine acoustic detail that allow dogs simply a result of living and interacting with the handler’s
to differentiate between 2 agonistic growls (Faragó et al., family, the better its performance and the relationship
2010). It is possible that dogs interpret low frequency human between dogs and handlers. Essentially, the more sociable
vocalizations, including warning tones, as potentially threat- dogs had better obedience performance than did less socia-
ening and high frequency human vocalizations as potentially ble dogs. Dogs that spent more time with handlers also
playful. However, given the likely importance of context in exhibited fewer of the stress-related behaviors that have
the interpretation of acoustic signals, it is probable that been used as indicators of welfare concerns in kenneled
the frequency of human vocalizations would be interpreted dogs, including pacing, barking, and destruction (Beerda
considering other cues and the history of the dog with those et al., 1999; Marston et al., 2004).
and similar human vocalizations. Conversely, it appears Horváth et al. (2008) studied responses to play sessions
that humans may tend to misinterpret the motivation behind between working dogs and their handlers and concluded
a canine vocalization as aggressive if it has low frequency that behaviors associated with control, authority, or aggres-
dispersion and thus sounds like a large dog (Taylor et al., sion increased cortisol concentrations, whereas play and
2010). It is possible there is a perceptual bias in humans to affiliative behavior decreased them. There is evidence that
perceive the growls of large dogs as more aggressive than cortisol can act as a hormone-response element and by
growls of small dogs (Taylor et al., 2009). doing so can stop the transcription of new proteins neces-
Humans can alter the types of vocalizations they give sary to make associative memory, an essential step in
to promote different behavioral responses from dogs. learning (Truss and Beato, 1993). The finding that handler
When humans send a signal consisting of short notes, it behavior can be associated with an increase in cortisol con-
can elicit a reactive response and increase motor activity centrations has profound implications for working dogs.
levels more than a signal consisting of a longer continuous Performance of Dutch police dogs trained with and
note (Yeon, 2007). Dalibard (2009) studied service dogs without shock has been compared (Schilder and van der
through a questionnaire survey and found that voice tone Borg, 2004). All dogs in both groups were successful police
and clarity affected how well and how quickly dogs dogs, but those in the unshocked group seemed to have a
responded to requests. smoother, more integrated relationship with their handlers,
Coutellier (2006) examined the responses of detection based on their response to their handlers’ signals.
dogs to the voice commands of their handler or a stranger. Positive and statistically significant associations
The response of the dogs differed significantly when the between obedience and the dog–handler relationship have
handler and handler’s voice were used, compared with the been found (Lefebvre et al., 2007). Handlers who interacted
stranger and stranger’s voice. There was no difference in with their dogs in a wider social context (e.g., took their
response if only the olfactory cues were changed, indicat- military working dog home or practiced sports with their
ing that dogs partially rely on acoustic information to dogs) reported more obedience and fewer bites from
perform their jobs. Despite the differences in the vocal their dogs. The same dogs also showed fewer behaviors
range of dogs and humans, there are some similarities associated with impaired welfare (pacing, barking, destroy-
that make some interspecific transfer of information plausi- ing items/materials), indicating that the effects of housing
ble. For example, it is possible that dogs pick up on the tone at a handler’s home and practicing sport were strongly
of loud vocalizations from us, allowing them to linked to an enhanced dog–handler relationship.
114 Journal of Veterinary Behavior, Vol 7, No 2, March/April 2012

Haverbeke et al. (2010) used the Human Familiarization to train a dog to walk next to a handler may provide insights
and Training Program (HFTP) to teach humans to under- into the handler’s ability to communicate effectively with the
stand normal canine signaling and to reward playful and dog. In this context, ‘‘communicate effectively’’ is meant to
appropriate behaviors. When compared with a control refer to the handler’s ability to train the dog to do the desired
group of dogs not participating in HFTP, the HFTP dogs task. For example, a dog can be trained to maintain a consis-
carried themselves higher, with less lowering of body pos- tent position relative to the handler when walking on the
tures associated with fear, showed less yawning, often a lead by strategically releasing the lead pressure when the
sign of uncertainty, and exhibited fewer aggressive behav- dog is in the correct position. It may also reflect subtle exam-
iors. These authors concluded that because of the changes ples of best practice in the way handler’s posture changes
in fearfulness, the welfare of HFTP dogs had improved. during heelwork (McGreevy, 2009).
Thus, here we see some exciting parallels between the
science of dog training and equitation science––the science
The way forward: measuring dog–dog of horse riding and training (McGreevy, 2007b). Capturing,
interactions and human–dog interactions defining, and measuring the qualities of techniques used by
the very best dog handlers are the essence of this approach.
The social relationship between dogs and humans has It seems that, despite the importance of on- and off-lead
been a topic of great interest in both the popular literature heelwork in the eyes of seasoned trainers, many owners use
(e.g., Serpell, 1995; Rogerson, 1988) and, more recently, the lead simply to restrain the dog and many dogs use the
scientific literature (e.g., Miklósi et al. 2004; Steinker, lead to control their owners. That said, we should not ignore
2007; Bradshaw et al., 2009). A social relationship may this critical interface. Even outside formal training, the aver-
be defined as regular and predictable behavior occurring age pet dog spends at least some time on the lead. Such
between 2 or more individuals (Scott and Fuller, 1965). periods are likely to be of enormous relevance to the dog be-
The 2 species interact behaviorally (McConnell, 2002) cause they represent time spent one-on-one with humans, and
and physiologically (Odendaal and Meintjes, 2003), but the endpoint (such as arrival at an exercise venue) is often
empirical studies of these interactions are only just emerg- highly reinforcing. In equitation science, rein tension meters
ing, despite their profound implications for the success of are an established means of measuring the mechanisms
individual relationships and the way in which lay people horse riders use to train horses with NR (i.e., the horse is
are educated to manage and train their dogs. Getting rewarded by the rider immediately releasing the tension in
this right can mean the difference between success and the rein when a horse performs the desired behavior)
failure in the human–dog interface. Rooney et al. (2000) (McGreevy and McLean, 2007). Using the same approach,
showed how intraspecific play styles of individual dogs a lead tension meter could be used to record the general level
were mirrored in those dogs’ play styles with humans. of contact, the intensity and frequency of all corrections a
The complexity of such human–animal interactions is handler applies through the lead, and the dog’s latency to re-
illustrated by recent studies of the temporal patterning of spond to them. Such data would capture the effectiveness
human–dog dyadic interactions (Kerepesi et al., 2005). In with which the handler uses NR or punishment and the extent
a similar vein, Jones and Josephs (2006) reported associa- to which the dog has become habituated to aversive stimuli
tions between salivary hormone concentrations in male from the collar.
dog handlers and their dogs. Specifically, dogs handled Considering principles beyond NR, equitation science
by men who underwent greater decreases in testosterone shows how operant conditioning of pressure cues can lay
concentration after losing a dog agility competition showed the foundation for the emergence of classical conditioned
greater increases in cortisol concentrations than in winning cues such as those from the rider’s seat or, most relevant
teams (Jones and Josephs, 2006). here, the handler’s posture. It explains how even without
Although defining social relationships as regular and leading a horse or applying physical pressure, hands-off
predictable behavior occurring between 2 or more individ- round pen techniques are effective in training horses. As
uals, Scott and Fuller (1965) also noted that they are neither such, it identifies some fascinating commonalities with the
invariable nor incapable of change. Although we agree with use of body language by dog trainers. The play-bow and the
this assessment of the fluid nature of the dog–human relation- body block (McConnell, 2002) are examples of postural
ship (changing with factors such as ontogeny, context, etc.), techniques that seasoned dogfolk use, often without being
we propose that quantifying certain features of the aware of what they are doing. We can use kinematic anal-
dog–human interaction may provide further insights that ysis systems that are emerging from equitation science to
increase our understanding of why some dog–human rela- study these techniques.
tionships are successful and others are not. Management Equitation science has identified the need for a working
factors have a significant impact on the success rate of dogs horse ethogram (Heleski et al., 2009). By the same token,
learning particular tasks (Batt et al., 2010) and in facilitating canine scientists need a validated canid intraspecific social
harmonious coexistence (McGreevy and Masters, 2008). For ethogram that exhaustively lists the types of behavior per-
example, the way that a particular tool, such as a lead, is used formed by domestic dogs in a social context (Plowman,
McGreevy et al Overview of the dog–human dyad 115

2010). It is accepted that many research groups have devel- Batt, L., Batt, M., Baguley, J., McGreevy, P.D., 2010. Relationships
oped their own ethograms but recent meetings of canine between puppy management practices and reported measures of suc-
cess in guide dog training. J. Vet. Behav.: Clin. Appl. Res. 5, 240-246.
scientists (such as the first and second Canine Science Bauer, E.B., Smuts, B.B., 2007. Cooperation and competition during
Forum in 2008 and 2010, respectively) serve to emphasize dyadic play in domestic dogs, Canis familiaris. Anim. Behav. 73,
the need for a canine ethogram that is validated to with- 489-499.
stand use by multiple observers and different laboratories Beerda, B., Schilder, M.B., Bernadina, W., van Hooff, J.A., de Vries, H.W.,
around the world. Lack of standardization will continue Mol, J.A., 1999. Chronic stress in dogs subjected to social and spatial
restriction: part II. Hormonal and immunological responses. Physiol.
to suppress the development of clear communication among Behav. 66, 243-254.
groups and obstruct attempts at meta-analysis (Tomkins Bradshaw, J.W., Blackwell, E.J., Casey, R.A., 2009. Dominance in domes-
et al., 2010). A validated ethogram will pave the way for tic dogsduseful construct or bad habit? J. Vet. Behav.: Clin. Appl.
the development of a related and cross-referenced dog– Res. 4, 135-144.
human ethogram. This tool is of critical importance because Coppinger, R., 2001. Dogs: A Startling New Understanding of Canine Or-
igin, Behavior and Evolution. Scribner, New York, NY.
with it canine scientists can advise veterinarians, handlers, Coutellier, L., 2006. Are dogs able to recognize their handler’s voice? A
trainers, and owners on the relative importance of subtle preliminary study. Anthrozöos 19, 278-284.
behavioral mechanisms operating at the dog–human inter- Dalibard, G.H., 2009. Parameters influencing service dogs’ quality of
face. Only then will we truly decipher the role, if any, of response to commands: retrospective study of 71 dogs. J. Vet. Behav.:
perceived social status in our dogs’ responses. Clin. Appl. Res. 4, 19-24.
Drews, C., 1993. The concept and definition of dominance in animal
behaviour. Behaviour 125, 283-313.
Conclusions Elgier, A.M., Jakovcevic, A., Barrera, G., Mustaca, A.E., Bentosela, M.,
2009. Communication between domestic dogs (Canis familiaris) and
It is critical that we study the way in which harmony is humans: dogs are good learners. Behav. Processes 81, 402-408.
Faragó, T., Pongrácz, P., Range, F., Virányi, Z., Miklósi, Á., 2010. ‘The
achieved among dogs and the extent to which communica- bone is mine’: affective and referential aspects of dog growls.
tion between dogs and humans can align with or breach the Anim. Behav. 79, 917-925.
framework offered by the dog–dog social ethogram. As- Feuerstein, N., Terkel, J., 2008. Interrelationships of dogs (Canis
pects of our body language and behavior may stimulate familiaris) and cats (Felis catus L.) living under the same roof.
resource-guarding and forms of anxiety in dogs. Displacing Appl. Anim. Behav. Sci. 113, 150-165.
Fiset, S., Gagnon, S., Beaulieu, C., 2000. Spatial encoding of hidden
dogs that have learned to defend their resources may objects in dogs (Canis familiaris). J. Comp. Psychol. 114, 315-324.
culminate in a bite. It is critical that we fully explore Fiset, S., Landry, F., Ouellette, M., 2006. Egocentric search for disappear-
possible mechanisms beyond a learning-theory framework ing objects in domestic dogs: evidence for a geometric hypothesis of
that may explain dog–human aggression and that the direction. Anim. Cogn. 9, 1-12.
emerging science of dog training continues to embrace Fiset, S., LeBlanc, V., 2007. Invisible displacement understanding in
domestic dogs (Canis familiaris): the role of visual cues in search
principles derived from both psychology and ethology. behavior. Anim. Cogn. 10, 211-224.
Fitch, W.T., 1997. Vocal tract length and formant frequency dispersion
correlate with body size in rhesus macaques. J. Acoust. Soc. Am.
Acknowledgments 102, 1213-1222.
Friedman, S.G., Brinker, B., 2001. The struggle for dominance: fact or
The authors thank Dr Wendy Birkhead for enlightening fiction? A bird’s eye view. Original Flying Machine 17-20.
Gácsi, M., McGreevy, P., Kara, E., Miklósi, Á, 2009a. Effects of selection
discussions on the problems faced by anthropologists when for cooperation and attention in dogs. Behav. Brain Funct. 5, 31.
grappling with the concept of ‘‘family.’’ Two anonymous Gácsi, M., Gyoöri, B., Virányi, Z., Kubinyi, E., Range, F., Belényi, B.,
reviewers provided excellent feedback on earlier drafts of Miklósi, Á, 2009b. Explaining dog wolf differences in utilizing human
this article. pointing gestures: selection for synergistic shifts in the development of
some social skills. PLoS One 4, e6584.
Goodwin, D., Bradshaw, J.W., Wickens, S.M., 1997. Paedomorphosis
affects agonistic visual signals of domestic dog. Anim. Behav. 53,
References 297-304.
Hahn, M.E., Wright, J.C., 1998. The influence of genes on social behavior
Ascione, F., 2008. The international handbook of animal abuse and of dogs. In: Grandin, T. (Ed.), Genetics and the Behavior of Domestic
cruelty: theory, research and application. Purdue University Press, Animals. Academic, New York, NY, pp. 299-318.
West Lafayette, IN. Haverbeke, A., Rzepa, C., Depiereux, E., Deroo, J., Giffroy, J.M.,
Brain, P.F., 2010. Aggression. In: Mills, D.S., Marchant-Forde, J.N., Diederich, C., 2010. Assessing efficiency of a Human Familiarisation
McGreevy, P.D., Morton, D.B., Nicol, C.J., Phillips, C.J., Sandoe, P., and Training Programme on fearfulness and aggressiveness of military
Swaisgood, R.R. (Eds.), Encyclopedia of Applied Animal Behaviour dogs. Appl. Anim. Behav. Sci. 123, 143-149.
and Welfare. CABI Publishing, Wallingford, Oxon, pp. 8-10. Heberlein, M., Turner, D.C., 2009. Dogs, Canis familiaris, find hidden
Barker, S.B., Rogers, C.S., Turner, J.W., Karpf, A.S., Suthers- food by observing and interacting with a conspecific. Anim. Behav.
McCabe, H.M., 2003. Benefits of interacting with companion animals. 78, 385-391.
Am. Behav. Sci. 47, 94-99. Heleski, C.R., McGreevy, P.D., Kaiser, L.J., Lavagnino, M., Tans, E.,
Batt, L., Batt, M., Baguley, J., McGreevy, P.D., 2009. The value of puppy Bello, N., Clayton, H.M., 2009. Effects on behaviour and rein tension
raisers’ assessments of potential guide dogs’ behavioral tendencies on horses ridden with/without martingales and rein inserts. Vet. J. 181,
and ability to graduate. Anthrozöos 22, 71-76. 56-62.
116 Journal of Veterinary Behavior, Vol 7, No 2, March/April 2012

Hennessy, M.B., Voith, V.L., Mazzei, S.J., Buttram, J., Miller, D.D., McGreevy, P.D., Calnon, D., 2010. Canine aggression in perspective. Vet.
Linden, F., 2001. Behavior and cortisol levels of dogs in a public animal J. 186, 1-2.
shelter, and an exploration of the ability of these measures to predict McGreevy, P.D., Grassi, T.D., Harman, A.M., 2004. A strong correlation
problem behavior after adoption. Appl. Anim. Behav. Sci. 73, 217-233. exists between the distribution of retinal ganglion cells and nose
Hennessy, M.B., Williams, M.T., Miller, D.D., Douglas, C.W., Voith, V.L., length in the dog. Brain Behav. Evol. 63, 13-22.
1998. Influence of male and female petters on plasma cortisol and be- McGreevy, P.D., Masters, A.M., 2008. Risk factors for separation related
haviour: can human interaction reduce the stress of dogs in a public distress and feed-related aggression in dogs: additional findings
animal shelter? Appl. Anim. Behav. Sci. 61, 63-77. from a survey of Australian dog owners. Appl. Anim. Behav. Sci.
Horowitz, A.C., 2009a. Attention to attention in domestic dog (Canis 109, 320-328.
familiaris) dyadic play. Anim. Cogn. 12, 107-118. McGreevy, P.D., McLean, A.N., 2007. The roles of learning theory and
Horowitz, A., 2009b. Disambiguating the "guilty look": Salient prompts to ethology in equitation. J. Vet. Behav.: Clin. Appl. Res. 2, 108-118.
a familiar dog behaviour. Behav. Processes 81, 447-452. McGreevy, P., Oddie, C., Burton, F., McLean, A., 2009. The horse-human
Horváth, Z., Dóka, A., Miklósi, Á, 2008. Affiliative and disciplinary dyad: can we align horse training and handling activities with the
behavior of human handlers during play with their dog affects cortisol equid social ethogram? Vet. J. 181, 12-18.
concentrations in opposite directions. Horm. Behav. 54, 107-114. McGreevy, P.D., Righetti, J., Thomson, P.C., 2005. The reinforcing value
Hsu, Y., Sun, L., 2010. Factors associated with aggressive responses in pet of physical contact and the effect on canine heart rate of grooming
dogs. Appl. Anim. Behav. Sci. 123, 108-123. in different anatomical areas. Anthrozöos 18, 236-244.
Hunt, M., Al-Awadi, H., Johnson, M., 2008. Psychological sequelae of pet McKinley, S., Young, R.J., 2003. The efficacy of the model-rival method
loss following Hurricane Katrina. Anthrozöos 21, 109-121. when compared with operant conditioning for training domestic
Jones, A.C., Josephs, R.A., 2006. Interspecies hormonal interactions dogs to perform a retrieval-selection task. Appl. Anim. Behav. Sci.
between man and the domestic dog (Canis familiaris). Horm. Behav. 81, 357-365.
50, 393-400. Mech, L.D., 1999. Alpha status, dominance, and division of labor in wolf
Kerepesi, A., Jonsson, G.K., Miklósi, Á, Topál, J., Csányi, V., packs. Canad. J. Zool. 77, 1196-1203.
Magnusson, M.S., 2005. Detection of temporal patterns in dog–human Miklósi, Á, 2007. Big thoughts in small brains? Dogs as a model for
interaction. Behav. Processes 70, 69-79. understanding human social cognition. Cogn. Neurosci. Neuropsy-
Kotrschal, K., Schöberl, I., Bauer, B., Thibeaut, A., Wedl, M., 2009. chol. 18, 467-471.
Dyadic relationships and operational performance of male and female Miklósi, Á, Topál, J., Csányi, V., 2004. Comparative social cognition: what
owners and their male dogs. Behav. Processes 81, 383-391. can dogs teach us? Anim. Behav. 67, 995-1004.
Lefebvre, D., Diederich, C., Delcourt, M., Giffroy, J.M., 2007. The quality Misra, R.P., 1995. Environmental Ethics: A Dialogue of Cultures. Concept
of the relation between handler and military dogs influences efficiency Publishing Company, New Delhi, India, p. 103.
and welfare of dogs. Appl. Anim. Behav. Sci. 104, 49-60. Moehlman, P.D., 1987. Social organization in jackals. Am. Sci. 75,
Ley, J., Bennett, P.C., Coleman, G.J., 2007. Personality dimensions that 366-375.
emerge in companion canines. Appl. Anim. Behav. Sci. 110, 305-317. Nicol, C., Pope, S., 1999. The effects of demonstrator social status and
Ligout, S., 2010. Social learning. In: Mills, D.S., Marchant-Forde, J.N., prior foraging success on social learning in laying hens. Anim. Behav.
McGreevy, P.D., Morton, D.B., Nicol, C.J., Phillips, C.J., Sandoe, P., 57, 163-171.
Swaisgood, R.R. (Eds.), Encyclopedia of Applied Animal Behaviour Odendaal, J.S., Meintjes, R.A., 2003. Neurophysiological correlates of
and Welfare. CABI Publishing, Wallingford, Oxon, p. 563. affiliative behaviour between humans and dogs. Vet. J. 165, 296-301.
Lit, L., Schweitzer, J.B., Oberbauer, A.M., 2010. Characterization of Overall, K.L., 1997. Clinical Behavioral Medicine for Small Animals,
human–dog social interaction using owner report. Behav. Processes Mosby, St. Louis, MO, pp. 11, 28.
84, 721-725. Pal, S., Ghosh, B., Roy, S., 1998. Agonistic behaviour of free-ranging dogs
Mack, A., 2003. Inattentional blindness: looking without seeing. Curr. Dir. (Canis familiaris) in relation to season, sex and age. Appl. Anim.
Psychol. Sci. 12, 180-184. Behav. Sci. 59, 331-348.
Marshall-Pescini, S., Valsecchi, P., Petak, I., Accorsi, P., Previde, E., 2008. Parker, G.A., 1974. Assessment strategy and the evolution of animal con-
Does training make you smarter? The effects of training on dogs’ flicts. J. Theor. Biol. 47, 223-243.
performance (Canis familiaris) in a problem-solving task. Behav. Petherick, J.C., 2010. Dominance. In: Mills, D.S., Marchant-Forde, J.N.,
Processes 78, 449-454. McGreevy, P.D., Morton, D.B., Nicol, C.J., Phillips, C.J., Sandoe, P.,
Marston, L.C., Bennett, P.C., Coleman, G.J., 2004. What happens to Swaisgood, R.R. (Eds.), Encyclopedia of Applied Animal Behaviour
shelter dogs? An analysis of data for 1 year from three Australian and Welfare. CABI Publishing, Wallingford, Oxon, pp. 185-186.
shelters. J. Appl. Anim. Welf. Sci. 7, 27-47. Plowman, A., 2010. Ethogram. In: Mills, D.S., Marchant-Forde, J.N.,
Masters, A., McGreevy, P., 2008. Dog keeping practices as reported by McGreevy, P.D., Morton, D.B., Nicol, C.J., Phillips, C.J.,
readers of an Australian dog enthusiast magazine. Aust. Vet. J. 86, 18-25. Sandoe, P., Swaisgood, R.R. (Eds.), Encyclopedia of Applied Animal
McConnell, P.B., 2002. The Other End of the Leash: Why We Do What Behaviour and Welfare. CABI Publishing, Wallingford, Oxon,
We Do Around Dogs. Bantam, Milsons Point, NSW. pp. 228-230.
McCrae, R.R., John, O.P., 1992. An introduction to the five-factor model Pongrácz, P., Molnár, C., Miklósi, Á, Csányi, V., 2005. Human listeners are
and its applications. J. Pers. 60, 175-215. able to classify dog (Canis familiaris) barks recorded in different
McGreevy, P.D., 2004. Equine Behavior–A Guide for Veterinarians and situations. J. Comp. Psychol. 119, 136-144.
Equine Scientists. W.B. Saunders, London. Pongrácz, P., Molnár, C., Miklósi, Á, 2006. Acoustic parameters of dog
McGreevy, P.D., 2007a. Breeding for quality of life. Anim. Welf. 16, barks carry emotional information for humans. Appl. Anim. Behav.
125-128. Sci. 100, 228-240.
McGreevy, P.D., 2007b. The advent of equitation science. Vet. J. 174, Pongrácz, P., Vida, V., Bánhegyi, P., Miklósi, Á, 2008. How does domi-
492-500. nance rank status affect individual and social learning performance
McGreevy, P.D., 2009. A modern dog’s life. UNSW Press, Randwick, NSW. in the dog (Canis familiaris)? Anim. Cogn. 11, 75-82.
McGreevy, P.D., Bennett, P.B., 2010. Challenges and paradoxes in the Pratto, F., Sidanius, J., Stallworth, L.M., Malle, B.F., 1994. Social domi-
companion animal niche. Darwinian selection, selective breeding nance orientation: a personality variable predicting social and political
and the welfare of animals. Anim. Welf. 19, 11-16. attitudes. J. Pers. Soc. Psychol. 67, 741-763.
McGreevy, P.D., Boakes, R.A., 2007. Carrots and Sticks: The Principles of Range, F., Heucke, S.L., Gruber, C., Konz, A., Huber, L., Virányi, Z.,
Animal Training. Cambrdige University Press, Cambridge, UK. 2009. The effect of ostensive cues on dogs’ performance in a
McGreevy et al Overview of the dog–human dyad 117

manipulative social learning task. Appl. Anim. Behav. Sci. 120, Szetei, V., Miklósi, Á, Topál, J., Csányi, V., 2003. When dogs seem to lose
170-178. their nose: an investigation on the use of visual and olfactory cues in
Ritvo, H., 1990. The Animal Estate. Penguin Books, New York, NY. communicative context between dog and owner. Appl. Anim. Behav.
Rogerson, J., 1988. Your Dog: Its Development, Behaviour and Training. Sci. 83, 141-152.
Popular Dogs Publishing, London. Taylor, A.M., Reby, D., McComb, K., 2009. Context-related variation in
Rooney, N.J., Bradshaw, J.W., Robinson, I.H., 2000. A comparison of the vocal growling behaviour of domestic dogs, Canis familiaris.
dog–dog and dog–human play behaviour. Appl. Anim. Behav. Sci. Ethology 115, 905-915.
66, 235-248. Taylor, A., Reby, D., McComb, K., 2010. Why do large dogs sound more
Schenkel, R., 1967. Submission; its features and function in the wolf and aggressive to human listeners: acoustic bases of motivational misattri-
dog. Am. Zool. 7, 319-329. butions. Ethology 116, 1155-1162.
Schilder, M.B., van der Borg, J.A., 2004. Training dogs with help of the Taylor, A.M., Reby, D., 2010. The contribution of source-filter theory to
shock collar: short and long-term behavioural effects. Appl. Anim. mammal vocal communication research. J. Zool. 280, 221-236.
Behav. Sci. 85, 319-334. Tomkins, L.M., McGreevy, P.D., Branson, N.J., 2010. Lack of standardiza-
Schwab, C., Huber, L., 2006. Obey or not obey? Dogs (Canis familiaris) tion in reporting motor laterality in the dog. J. Vet. Behav.: Clin. Appl.
behave differently in response to attentional states of their owners. Res. 5, 235-239.
J. Comp. Psychol. 120, 169-175. Truss, M., Beato, M., 1993. Steroid hormone receptors: interaction with
Scott, J., Fuller, J., 1965. Genetics and the Social Behavior of the Dog. deoxyribonucleic acid and transcription factors. Endocr. Rev. 14,
University of Chicago Press, Chicago, IL. 459-479.
Semyonova, A., 2003. The social organization of the domestic dog; a Tuber, D.S., Hennessy, M.B., Sanders, S., Miller, J.A., 1996. Behavioral
longitudinal study of domestic canine behavior and the ontogeny of and glucocorticoid responses of adult domestic dogs (Canis familiaris)
domestic canine social systems. The Carriage House Foundation, The to companionship and social separation. J. Comp. Psychol. 110, 103-108.
Hague, Netherland. Version 2006. Available at: www.nonlineardogs. van Kerkhove, W., 2004a. A fresh look at the wolf-pack theory of
com. Accessed October 15, 2010. companion-animal dog social behavior. J. Appl. Anim. Welf. Sci. 7,
Serpell, J., 1995. The Domestic Dog: I Evolution, Behaviour, and Interac- 279-285.
tions With People. Cambridge University Press, Cambridge, UK, van Kerkhove, W., 2004b. Response to comments. J. Appl. Anim. Welf.
New York, NY. Sci. 7, 299-300.
Shepherd, K., 2002. Development of behaviour, social behaviour and com- Vas, J., Topál, J., Gácsi, M., Miklósi, Á, Csányi, V., 2005. A friend or an
munication in dogs. In: Horwitz, D.F., Mills, D.S., Heath, S. (Eds.), enemy? Dogs’ reaction to an unfamiliar person showing behavioural
BSAVA Manual of Canine and Feline Behavioural Medicine. BSAVA, cues of threat and friendliness at different times. Appl. Anim. Behav.
Quedgeley, UK, pp. 8-20. Sci. 94, 99-115.
Sheppard, G., Mills, D.S., 2002. The development of a psychometric scale Voith, V.L., 2009. The impact of companion animal problems on society
for the evaluation of the emotional predispositions of pet dogs. Int. and the role of veterinarians. Vet. Clin. North Am. Small Anim. Pract.
J. Comp. Psychol. 15, 201-222. 39, 327-345.
Sherman, C., Reisner, I., Taliaferro, L., Houpt, K., 1996. Characteristics, Wilsson, E., Sundgren, P.E., 1998. Behaviour test for eight-week-old
treatment, and outcome of 99 cases of aggression between dogs. puppies-heritabilities of tested behaviour traits and its correspondence
Appl. Anim. Behav. Sci. 47, 91-108. to later behaviour. Appl. Anim. Behav. Sci. 58, 151-162.
Slabbert, J.M., Rasa, O.A., 1997. Observational learning of an acquired Wright, J.C., 2004. Comment on van Kerkhove’s "Wolf-Pack Theory".
maternal behaviour pattern by working dog pups: an alternative train- J. Appl. Anim. Welf. Sci. 7, 295-298.
ing method? Appl. Anim. Behav. Sci. 53, 309-316. Wrubel, K.M., Moon-Fanelli, A., Maranda, L., Dodman, N., 2011. Inter-
Soproni, K., Miklósi, Á, Topál, J., Csányi, V., 2001. Comprehension of dog household aggression: 38 cases (2006–2007). J. Am. Vet. Med.
human communicative signs in pet dogs (Canis familiaris). J. Comp. Assoc. 238, 731-740.
Psychol. 115, 122-126. Yeon, S.C., 2007. The vocal communication of canines. J. Vet. Behav.:
Soproni, K., Miklósi, Á, Topál, J., Csányi, V., 2002. Dogs’ (Canis Clin. Appl. Res. 2, 141-144.
familiaris) responsiveness to human pointing gestures. J. Comp. Yin, S., 2002. A new perspective on barking in dogs (Canis familiaris).
Psychol. 116, 27-34. J. Comp. Psychol. 116, 189-193.
Steinker, A., 2007. Terminology Think Tank: Social dominance theory as Yin, S., 2009. Low Stress Handling, Restraint and Behaviour Modification
it relates to dogs. J. Vet. Behav.: Clin. Appl. Res. 2, 137-140. of Dogs and Cats. Cattle Dog Publishing, Davis, CA.